iPod nano/mini 2G use clock speed optimisation to increase battery time.
VBR songs with a sudden increase in bitrate may cause the iPod nano/mini 2G to stutter as the clock speed is not increased fast enough to deal with the increased bitrate.

This has so far been observed for songs encoded with LAME --preset standard and is only true when not using the equaliser.

Examples of songs that stutters can be found here:
David Bowie - Five Years
Nine Inch Nails - Right Where it Belongs

For Five Years the clock speed hovers around 24 jumping to 30 when the stuttering begins. Using the equaliser the clock speed hovers around 36 throughout the song and no stuttering occurs.

Further observations:
- this problem is related to the iPod nano/mini 2G but one sample has been found that makes the 4G Color iPod stutter
- problem with iPod mini software 1.3, 1.4, and 1.4.1
- problem with iPod nano software 1.0, and 1.1
- iPod mini 2G do not use optimisation when connected to a computer
- iPod mini 2G do use optimisation, both when using its battery and when connected to a powered dock that is not connected to a computer
- problem go away when using an EQ setting different from flat/off
- only safe EQ setting to use is Bass Reducer / Treble Reducer in order to avoid distortion of songs with a high recording level
- problem do not occur when using iTunes MP3 encoder
- problem do not occur when LAME VBR song do not contain 320 kbps: options --preset standard -B 256
- problem do not always occur at the same spot in the subtle songs listed below
- iPod mini 1G clock speed for Five Years hoovers aroung 31 rising to 33 where it was 24 / 30 for the 2G
- iPod mini optimisation may be turned off (I don't now how to provoke this) and require a reset of the iPod to turn it back on.
- for the 4G color iPod only the sample with NIN - Right Where it Belongs stutters, even with the EQ set to Bass Reducer. Acoustic EQ setting makes the problem go away.

Songs having the problem (obvious):
- Johnny Cash - A Satisfied Mind
- David Bowie - Five Years
- Nine Inch Nails - Right Where it Belongs
- Cake - Nugget @ 26s

Songs having the problem (subtle):
- Eminem - Drug Ballad
- Eminem - Kill You
- U2 - Miracle Drug
- Musica XM - AWS Hope Xm cut

Album songs not having the problem:
- Pink Floyd - Dark Side of the Moon
- Styx - Pieces of Eight
- Guns N' Roses - Illusions I
- Shirley Bassey - Greatest Hits/This is my Life

Edited: Updated list of songs and albums.
Edited: iPod mini 1G do use speed optimisation. My iPod mini 1G was stuck at a clock speed of 80 and required a reset forcing it to use speed optimisation.
Edited: One user with iPod mini 2G was unable to reproduce problem.
Edited: Work around described
Edited: Added NIN
Edited: Added Cake
Edited: Updated for iPod mini software 1.4
Edited: Added Johnny Cash and updated observations
Edited: Added information about 4G color iPod stuttering
Edited: iPod nano stutters as bad as iPod mini 2G
Edited: Updated for iPod mini software 1.4.1 / iPod nano software 1.1

Wow... that's kind of a big bug.

Should be reasonably easy for Apple to fix with a firmware update though.

I've got a 1G iPod mini, and no problems at all for me...

All new iPods use variable clock speed, but there were some speculation that the mini uses the brand new ARM chip that haven't officially gone into mass production yet. But yeah, it should be easy to fix if they are aware of it.

all my VBR tracks play fine on my (technically 2G) 30GB ipod photo... except some on Nine Inch Nails - With Teeth. the 192k worked flawlessly, but the VBR rip stutters every once in a while. I thought maybe it was an Xfering problem, but I re-upped it several times and the problem still persisted. . . interesting . and it was encoded with lame 3.90.3



that , and the original mp3s are flawless -- no skips on 'puter playback

I have provided feedback to Apple and posted it on iPodLounge, so eventually I hope Apple will become aware of the problem and fix it.

How much of a problem it is is not yet clear as so far only LAME aps have been tested.

It might be that I have just scratched the surface and the problem turns bigger when investigated.

Anyway it is clearly a trade-off between battery time and minimising the risk for stuttering. In the current marketing war it might be that Apple will prioritise outstanding battery performance over occasional stuttering of a few songs, as long as the songs are not in AAC.

Have you checked if you can reproduce with VBR files encoded with iTunes?

No. Anyway lame aps is the dominant format also used by several download services.

If I where to use an iTunes specific format I would chose AAC.

However I have checked ALAC, which is also a VBR format, and there was no stuttering.

My priority now will be to listen to music in lame aps to identify the magnitude of the problem.

Yea, but if it happens with their own encoder as well, it would probably get a higher priority.

that's wrong. I own an 1G iPod mini and the clock speed jumps up and down as well while playing. If you start to listen or skip a song it will jump to 80 for a few seconds.

Anyway, couldn't you just prepare a sample...let's say digital silence followed by some part that will make LAME to use 320kbps? Maybe --preset extreme could help to create such a scenario...

That's interesting. I would have expected the same, but the iPod mini 1G I tested did not change from 80 on the sample provided. Software version was 1.3.
What would the point be? I already provided a sample that demonstrates the problem.

I have a 4G 20GB iPod and I'm pretty sure its processors change speed dynamically during playback (see this article), but I've never once encountered a stutter in the 5 months I've owned it. Almost all of my 1,500+ song library is encoded in LAME -preset-extreme via EAC.

try to reset your iPod using the ipod_updater software...or you could also (un)-cap your iPod...this makes the iPod loading the defaults, which should fix the problem. I sometimes have that problem too - and I'm using the latest firmware as well. I usally notice it when my battery is already drained after 4 hours...

nope your iPod should do fine as well as the 1G iPod mini (even though they both shift clock speeds). There seems to be a problem with the new chips Apple uses for their recent models that need less CPU power and therefore result a longer playback time...as far as I know only 2G iPod mini and 2G iPod Photo use these new chips...

My 2G 4GB mini is filled with --aps MP3s (ripped using EAC's secure mode) and I haven't had any problems with stuttering. I listen to all kinds of different music too. I have everything from Alicia Keys to Fear Factory on my iPod, so theres a variety of complexity and types of music. Not a single song has skipped or stuttered.

I downloaded the sample posted in the other thread. So far I've listened to it about 6 times and it's played flawlessly every time. I sent it over to my friend with a 6GB 2G mini and hes had no problems with it either.

I hate to say this, but I think it's just a problem with the original poster's iPod. So far I've seen 2 iPods play it correctly that, according to the original poster, should have stuttered at least once.

my 2G ipod mini skips with 192k cbr songs too.. very randomly also.. so I don't think it's just the bitrate.. some frames might be harder to decode than others even if they're the same size, right? if not, I don't get it.

Where do you check the clock speed?

edited: typo

Most songs seems not to be affected by this problem. I'm updating the original post as I found new examples.


It's not just one iPod mini 2G that has the problem: HelloWhat

Are you sure clock speed optimisation was in use when you played back the sample?

To view the clock speed of an iPod (Windows version) do the following:
Connect iPod to your computer
Launch Windows Explorer
Change folder settings to show hidden files and folders. Also don't hide file type of known files
Create an empty file named "_show_speed" in iPods iPod_Control\Device folder
Picture of Windows Explorer displaying show speed configuration
Disconnect the iPod
Set Date & Time setting "Time in title" to On


Side Note: Is there a simple way to place pictures in posts, when you don't have your own web server?

Thanks for the advice.

After launching iPod Updater and reseting the iPod speed optimisation came into effect.

I have updated my original post to reflect this new information.

I spoke too fast.. I transfered one of the (many) songs that skip on my iPod, checked it with winamp, and it was a VBR file.. it was encoded my me, so it could only be -aps. I thought these tracks were from another much older 'batch' I encoded at 192k cbr.. seems I was wrong

The point is I just checked again and it skips. Thanks for that _show_speed thing, very useful when playing these files the cpu varies from 27 to 30.

I've found a part of a song that causes my iPod to skip. Always does it, it isn't random. I've uploaded it here:

http://www.hydrogenaudio.org/forums/index....71&#entry293071

Clock speed optimisation was on. Infact, I even reset it just to make sure. I've been listening to both songs posted in the other thread for about the last 10 minutes or so. Constant repeat. No stuttering whatsoever.

The clockspeed generally hung around 22-28MHz for both songs. Strangely enough, I tried a 128kbps AAC song, and the clockspeed is higher at around 28-30MHz. hmm.

Anyway, no stuttering for me.

So not all 2G mini's will stutter. I will update my original post to this new information.
It is commonly known that AAC use more juice than MP3 to decode:
MP3 or AAC for iPod Battery Life

i was listening to ipod photo last nite, and it skipped on a weezer song. i replayed it , and it skipped exactly where i thought it did before. and it was a flawless rip -- again, no skips on 'puter playback. this must be fiXed!

Please provide direct feedback to Apple. The more complaints, the higher priority they will give the issue.

this sounds to me like a simple buffering problem - like what you'd get with winamp running on a pentium 100 while something else was running (hehhe... oldskoo). if you increase the buffering, the file played back smoothly.

i wonder if that's controllable on the pod? i don't own one and probably never will (they're too fashionable for a geek like me to have in good conscience ) unless apple implement gapless playback and perhaps PCM recording to give a use to those big hard disks they have.

i just used the URL u gave me and told apple... hopefully a fix will be out soon, hopefully they care, and hopefully its not a hardware problem.

So much for Apple's claims that the I-Pod supports MP3-VBR.

i wonder why we are the first to complain.. what are you going to do about it?

edit - i think i am going to transcode to 128 AAC

i think downgrading from 1.1 to 1.0 fixes the problem... i will report back once i've had extensive testing

i havent given it *insane* testing, but try rolling back your software to that of 2005-1-11 if possible with 2G minis... did the trick with my ipod photo.. at least it doesnt skip where it did before. will report back after days of listening

EDIT - nevermind i heard it skip, this really sucks, i think i might just sell it or take it back if theres no fix soon

EDIT 2- i'm sending it back and hopefully the replacement will fix it.

is this not an issue with 1G ipod photo 40GB and 60GB?

if so thats bizarre, because both 1G and 2G afaik clock at 15 hours. right?

I'm encountering exactly the same problems with my iPod Mini 2G 6 GB.
I also have the stutter problem and I did an firmware update, hoping that would solve the problem. After updating I tested the Mini in the dock (which I normally only use for connecting the iPod to the PC for updating my music libary) with some 'problem' mp3's ... no stuttering any more .... I thought ... because after that the stuttering was 'reintroduced' while travelling by train (when the iPod is playing from battery).

Very complex and dynamic music (like Tool ) stutters quite often and that really sucks, because I like my iPod very much and I think there isn't any good alternative for an iPod (flashplayers, Creative, etc. ... yuk ), but an iPod with this problem also isn't heaven

I've just filled in the form 'iPod Feedback' and i really hope Apple is gonna solve this problem, because it's very annoying.

It's not specific to the Mini.

I get the same problem with my 40GB 4G iPod, as do several people on a forum I frequent.

One thing one of them noticed...if you use the Rock EQ preset...the stutters go away. Other EQs do not have this effect. Can someone try it with their Mini?

That works!

When using an EQ setting other than flat/off the clock speed increases to 35 and stuttering is no longer a problem.

That's a good work around while waiting for Apple to sort out their problem.

Only downside is that it will be a bit more battery hungry.

I have updated the original post to reflect this new information.

cool thanks, only problem is that i returned my ipod to apple for repair and they said it passed the diagnostics, thus they didn't do anything to fix it.

DARN APPLE DARN

this is cool that it works with rock eq, but sort of lame too. i dont really like this equalizer. but if its the only way....

So wait...does this indicate that there is a processor difference between the 20GB and 40GB 4G iPods? As I mentioned previously, I have a 20GB 4G iPod, nearly my entire library is in LAME 3.96.1 -present extreme (some standard), and I do most of my listening with the equalizer set to "Off" (especially when playing through my main stereo system)...and I have never once experienced the stutter you all described.

hmmm, any way to fix this besides using the equalizers?

this really really is beginning to piss me off. it has to be a software problem , right?

EDIT- sehested , is there any thing you can do ? you are pretty good with ipods.

The only two work arounds I know of are:
- Use equaliser
- Transcode to AAC

Eventually I beleive Apple will come with a software update that will fix the problem.

However until then I meditate and pray that the force will be strong in me - allowing me to keep my cool.

yeah if it isn't fixed in the next software update, i will raise hell (i.e sell my photo, and probably just buya 1G 60GB for a bit more; cross my fingers that one wont have the problem either)

If making fresh vbr mp3s, I see some options which might help,
Latest versions of Lame can use 128 as minimum bitrate for some presets, but adding -b 128 to be sure and -F to stop it dropping to 32Kbs for silences could help.
- This issue is surely to do with power management and 32 kbs frames at the start of a track could be misleading the decoder.
Quality should not be hurt by this, and although the files will be bloated slightly, the bit reservoir could recycle alot of the spare bits.

Also at high bitrates like 192 and 224, 256, vbrs advantage over cbr becomes very marginal imbo.

When using David Bowie - Five Years the test results are:
lame --preset standard : 165kbps - clock speed 24 - Obvious stuttering
lame --preset standard -b 128 -F : 169kbps - clock speed 24 - Obvious stuttering
lame --preset standard -b 160 -F : 180kbps - clock speed 24 - Stuttering
lame --preset standard -b 192 -F : 198kbps - clock speed 24 - Stuttering
lame -b 192 : 192kbps - clock speed 24 - No stuttering
lame -b 256 : 256kbps - clock speed 28 - No stuttering
lame -b 320 : 320kbps - clock speed 30 - No stuttering
lame --preset standard -B 192 : 150kbps - clock speed 24 - No stuttering
lame --preset standard -B 256 : 160kbps - clock speed 24 - No stuttering

Conclusion:
There must be a jump to 320 kbps in the VBR that the iPod can't handle by increasing its clock speed fast enough from 24.

You do seem to have located the problem, Im surprised the -b 128 -F had little effect. It certainly Looks like Apples' bug rather than Lames unexpected agility.
Apple or someone who can dissasemble the firmware could fix this with a 'single byte hack' to up the power mode at 256 to 320's speed, with just a tiny hit in power drain...

Edit:
or does it look like, for vbrs the clock speed is static at 24 whatever the the range of bitrates :puzzled: Maybe the readout is not to be trusted, because there is a small observed improvement in > -b 160 but no observed clock speedup.

I got annoyed by the skips. selling my 30gb with case for 325 shipped to someone...

great for me, i only lost 30 bucks after a month of usage.

it was 334 shipped with educational discount and 20 for case. so 354 total originally.


and yes i told them about the glitch, buyer doesnt care, he rips with iTUNES anyways

contemplating 1G ipod photo 40GB/60GB or 2G mini (that has been thoroughly inspected and doesnt have the skip) 4GB/6GB

You can be pretty confident Apple will fix it soon, as issues like this have occurred before and has been remedied by a firmware update. You also know that it dissapears with the EQ turned on, right?

In any case, it's not something that happens to just a few models. All 4G's, photo's and 2G minis seem to have it

This is nonsense. The 2G mini stutters for sure.

it is truly ALL of them ? i dont think so. as stated before by a 4G user, he has never had a skip. its a software problem, but there's just quite a few unlucky units that have the stinkware in full effect.

oh well, either way, I want a 1G photo . axxories are the way to go; it really adds to the product's value and I can get a 60GB sealed for 393 shipped (Not ebay either)

edit- if im wrong, and all the versions stated skip with vbr, i'll sit on the purchase until (if) it is fixed.

i sent my ipod in stating the problem, and they said there was no problem. the techs didnt even acknowledge it. at the very least they could've said apple is aware and will be fixing it with software, but this isn't enough reason to give you a new unit.

but noooo, none of that.

just loaded up Nine Inch Nails - Right Where It Belongs (towards end is the major skip-test) on my dad's 2G 4GB mini, and it skipped

if no fix comes out in the next 2 months, i'm going to buy a non-apple DAP. this is unacceptable and i'm getting impatient (although my 3G 15GB is suffice)

I remember when the 3G iPods were released, a loud clicking sound happened between each song and the left/right channel swapped each time you hit pause. A couple of months later and a firmware update fixed it completely. The same thing will probably happen here as well (as long as they know about it)

I'm not really sure what is the most effective way to convey the problem to Apple.

The three "official" channels to report problems, that I know of are:
- Direct feedback through the Apple web site
- Posting the problem on Apple's discussion forums
- Contacting support by phone and demanding a solution, insisting them to report the problem to their second line support.

The problem with all three are that Apple do not provide any feedback or acceptance of the reported problem.


My personal experience with Apple support:

When I contacted Apple support by phone, they didn't know what LAME was. Their reaction was a shoulder shug as he suggestion to use the built-in mp3 encoder. After the supporter had conferred with someone else, he suggested that I visited iPodLounge or Apple discussion forums to look for a solution.

I informed that I was probably the first to discover the problem and had already posted on both forums. I suggested that he could get full details about the problem by searching for stuttering on these forums. I insisted, arguing that LAME has been used to encode the majority of mp3 files floating around in the world, that Apple should provide a software update that will fix the problem.

The supporter had to discuss the problem yet again with a collegue / superior. He promised to bring it to the developers attention and informed me that the developers do visit the Apple Discussion forums as well as gets the Bug report feedback.

The call took altogether about 30 minutes.

The only feedback I got was a case number...


Anyway I still beleive Apple will eventually fix this problem, so I can't wait for the annoncement of the next update.

@Busemann: I sure hope the right people know about it...